You are on page 1of 1

1 What is Deep Learning?

In simple words, Deep Learning can be understood as an algorithm which is composed


of hidden layers of multiple neural networks. It works on unsupervised data and is
known to provide accurate results than traditional ML algorithms.
Input data is passed through this algorithm, which is then passed through several nonlinearities before delivering output. This algorithm allows us to go deeper (higher level
of abstraction) in the network without ending up writing lot of duplicated code, unlike
shallow algorithms. As it goes deeper and deeper, it filter the complex features and
combines with those of previous layer, thus better results.
Algorithms like Decision Trees, SVM, Naive Bayes are shallow algorithm. These
involves writing lot of duplicated code and cause trouble reusing previous computations.
Deep Learning through Neural Network and takes us a step closer to Artificial
Intelligence.

2 What is Neural Network?


The concept of Neural Network began way back in 1980s. But, has gained re-ignited
interest in recent times. Neural network is originally a biological phenomenon. Neural
network is a network of interconnected neurons which maintain a high level of
coordination to receive and transmit messages to brain & spinal cord. In machine
learning, we refer Neural Network as Artificial Neural Network.
Artificial Neural Network, as the name suggests, is a network (layer) of artificially
created neurons which are then taught to adapt cognitive skills to function like human
brain. Image Recognition, Voice Recognition, Soft Sensors, Anomaly detection, Time
Series Predictions etc are all applications of ANN.

You might also like